Network Business And The Www Cache Technology

The World Wide Web is based on Hypertext Transfer Protocol, which works as Client/Server model. When it is used for information distribution, the C/S model is not adaptable to scalability and the bandwidth is wasted greatly. Caching popular objects is the key technology to alleviate Web server bottleneck, reduce traffic over the Internet and reduce response time. A lot of work has been done in this area since Web appears. This paper discusses in depth the key technology of web caching to improve WWW performance.The main contributions of this paper are as follows. A statistical multiplexing method of MMBP source inputs is presented. We summarize the currently often used traffic models. Using MMBP inputs, we analyze the performance of multiplexer, which can be applied to voice and data modeling. The singularity of matrix is avoided with the matrix analysis method, which exaggerates the queue average length as in literatures. Experiment verifies this point. Based on the analysis of Web traces, we present an adaptive replacement policy. The policy considers each important aspect that affects replacement effectivity, including reference, recent access time, document size, cost and user access pattern. The experiment shows the new replacement policy is better than current policies in literatures. A new cooperative caching system is presented. By distributing URL space evenly and randomly, heavy overhead caused by communication between proxies is avoided. Using proxy manager to distribute URL requests, the system really becomes a virtual, large cache as a whole. The new system also overcomes the defect of high redundancy among proxies. A content consistency strategy of the hierarchical cooperative web caching system is provided. The function of the existing Time-to-Live method and Lease method is extended and applied to our system.